Microsoft has unveiled two new features for its Copilot Researcher, Critique and Council, integrating OpenAI's GPT with Anthropic's Claude to enhance research capabilities. Critique uses a sequential collaboration model where GPT manages research planning and draft creation, while Claude ensures factual accuracy and citation quality. Council allows both models to independently generate reports, with a third model synthesizing their outputs. In the DRACO benchmark, Copilot with Critique achieved a score of 57.4, outperforming the second-place model by nearly 14% and significantly exceeding Claude Opus 4.6's score of 42.7.